Receiving streaming content from servers located around the globe

1. A method for retrieving streaming content by an assembling device, comprising: pulling via the Internet, by the assembling device from a plurality of fractional-storage servers, a set of erasure-coded fragments associated with segments of streaming content, each fragment pull request associated with a specific segment; wherein the fractional-storage servers are scattered over at least one continent and operative to have an aggregated fragment delivery bandwidth greater than the incoming bandwidth of the assembling device; and each of the fractional-storage servers is configured to store at least one erasure-coded fragment but less than the set of erasure-coded fragments; approaching the incoming bandwidth of the assembling device by pulling the fragments fast enough and essentially regardless of fragment loss or distances between the assembling device and the servers; receiving, by the assembling device, at least some of the requested fragments; and compensating for lost fragments by obtaining by the assembling device additional erasure-coded fragments that are needed to reconstruct the segments.

2. The method of claim 1 , wherein the step of pulling the fragments utilizes a fragment pull protocol, the fragment pull protocol is a fragment pull protocol for high latency, and the step of compensating for the lost fragments is performed essentially independently of the rate of pulling the fragments.

3. The method of claim 1 , wherein the step of pulling the fragments utilizes a fragment pull protocol, the fragment pull protocol is a fragment pull protocol for high latency, and the step of requesting the fragments is performed at a rate that is essentially independent of the rate of the fragment loss.

4. The method of claim 1 , wherein the step of pulling the fragments utilizes a fragment pull protocol, the fragment pull protocol is a fragment pull protocol for high latency, and controlling download rate of the streaming content using the rate of requesting the fragments.

5. The method of claim 1 , wherein the step of pulling the fragments utilizes a fragment pull protocol, the fragment pull protocol is a fragment pull protocol for high latency, the streaming content is high-definition streaming media, and the erasure-coding is rateless-coding.

6. The method of claim 1 , wherein the step of pulling the fragments comprises requesting from fractional-storage servers having an average one-way network-related latency greater than 50 milliseconds between the assembling device and the fractional-storage servers, and the erasure-coded fragments are rateless-coded fragments.

7. The method of claim 6 , wherein the step of compensating for lost fragments comprises obtaining the additional fragments from the same plurality of servers.

8. The method of claim 6 , wherein the step of compensating for lost fragments comprises obtaining the additional fragments from a different server.

9. The method of claim 6 , wherein the step of compensating for lost fragments comprises obtaining the additional fragments from a low latency server.

10. The method of claim 1 , wherein the fractional-storage servers are fractional-storage CDN servers located close to or on the Internet backbone and configured to have a high storage gain.

11. The method of claim 1 , further comprising sending, by the assembling device, multiple fragment pull requests over a period shorted than the average round-trip network-related assembling-device-servers latency in order to reach fragment delivery throughput approaching the available incoming bandwidth of the assembling device within a period shorter than two times the average round-trip network-related assembling-device-servers latency.

12. The method of claim 1 , wherein the step of pulling the fragments utilizes a fragment pull protocol, the fragment pull protocol is a fragment pull protocol for high latency, and further comprising categorizing the servers into fastest responding servers, and slower responding servers; pulling more than half of the fragments using the fragment pull protocol for high latency from the slower responding servers, and the step of compensating for lost fragments comprises obtaining the additional erasure-coded fragments quickly from the fastest responding servers.

13. A system comprising: a plurality of fractional-storage servers, scattered over at least one continent, configured to store erasure-coded fragments associated with segments of streaming content; each of the fractional-storage servers is configured to store at least one erasure-coded fragment but less than a certain set of the erasure-coded fragments; and an assembling device configured to request and receive, using a fragment pull protocol over the Internet, from the fractional-storage servers, the set of the erasure-coded fragments, each fragment pull request associated with a specific segment, and compensate for lost fragments by requesting additional erasure-coded fragments, wherein the servers operative to deliver fragments at a bandwidth greater than the incoming bandwidth of the assembling device, and the streaming-content-retrieval bandwidth of the assembling device is configured to approach the incoming bandwidth of the assembling device, by requesting the fragments fast enough and essentially regardless of fragment loss or distances between the assembling device and the servers.

14. The system of claim 13 , wherein the fractional-storage servers are fractional-storage CDN servers located close to or on the Internet backbone, the fragment pull protocol is a fragment pull protocol for high latency, and the erasure-coding is rateless-coding.

15. A system comprising: fractional-storage servers configured to store erasure-coded fragments associated with segments of streaming content; each server configured to store at least one erasure-coded fragment but less than a first set of the erasure-coded fragments; and an assembling device operative to have an average one-way network-related latency of more than 50 milliseconds to the fractional-storge servers, the assembling device configured to obtain the first set of fragments from the fractional-storage servers ,and to compensate for lost fragments by obtaining a second set of erasure-coded fragments usable to reconstruct the segments, the fractional-storage operative to deliver fragments at an aggregated bandwidth greater than the incoming bandwidth of the assembling device, and the assembling device is further configured to obtain the fragments at a bandwidth approaching the incoming bandwidth of the assembling device, essentially regardless of fragment loss.

16. The system of claim 15 , wherein the assembling device is configured to utilize a fragment pull protocol to obtain the fragments.

17. The system of claim 16 , further comprising at least one server located relatively close to the assembling device; wherein the assembling device is configured to obtain a percentage of the second set of fragments from the close server.

18. The system of claim 16 , wherein the assembling device is further configured to categorize the fractional-storage serves into at least two categories comprising fastest responding fractional-storage servers, and slower responding fractional-storage servers; and configured to essentially avoid obtaining the first set fragments from the fastest responding fractional-storage servers, and obtain at least most of the second set fragments quickly from the fastest responding fractional-storage servers.

19. The system of claim 15 , wherein the fractional-storage servers are located close to or on the Internet backbone, and the erasure-coding is rateless-coding.

20. The system of claim 15 , wherein the assembling device is further configured to obtain the fragments using multiple streams from multiple fractional-storage servers,each stream contains less than half of the required fragments.